The Decomposition of p-Methylbenzyl Hydroperoxide I

4b Lorand and Edwards observed ... J. Durham, C. F. Wurster and H. S. Mosher, ibid., 80, ..... (16) S. Winstein, B. K. Morse, E. Grunwald, K. C. Schre...
0 downloads 0 Views 760KB Size
WILLIAMJ. FARRISSEY, JR.

1002 [CONTRIBUTION FROM

THE

Vol. 84

RESEARCH AND DEVELOPMENT DIVISION,HUMBLEOIL

AND

REFININGC O . , BAYTOWN, TEX.]

The Decomposition of p-Methylbenzyl Hydroperoxide BY WILLIAM J. FARRISSEY, JR. RECEIVED SEPTEMBER 25, 1961 The thermal decomposition of p-methylbenzyl hydroperoxide ( I ) has been examined in several solvents. Decomposition vie a hydroxyperoxide (111) is much less important than observed for primary aliphatic hydroperoxides. In the presence of n-heptaldehyde, the hydroxyperoxide IV becomes the major pathway for decomposition, particularly a t lower ternperatures. The degree of participation of I11 as a decomposition intermediate can be increased by acids, p-tolualdehyde and p-cresol. Decomposition in acetic acid results in formation of p-cresoxymethyl acetate (V). OH Introduction I a It is generally accepted that hydroperoxides RCHzOOCH-R‘ +RCHO R’COOH Ht (1) play an important role in the autoxidation of hydrocarbons. l s 2 Their intermediacy has been demonRCH2OCOR’ H20 strated in a variety of autoxidation systems.2 It 11. R = R’ = alkvl is not surprising, then, that considerable work has been reported on the chemistry of hydroperoxides of various types. One important class of hydroperoxides, however, has received scant attention ; namely, the primary benzyl hydroperoxides. Only tibility of aromatic aldehydes to nucleophilic attwo examples of this class are reported in the liter- tack,6 however, may retard formation of I1 to an extent that i t does not contribute greatly to the ature, benzyl3 and p - m e t h y l b e n ~ y l ~hydroper~,~ oxides, of which only the latter has been examined over-all decomposition reaction. This and other aspects of the thermal decomposition of I are the with regard to its decomposition behavior. The synthesis of p-methylbenzyl hydroperoxide objects of this study. The synthesis of I was achieved essentially as (I) was first reported by Hock and Lang,4a who After sepadetermined that, in common with secondary described by Lorand and ration from the oxidation mixture by extraction hydroperoxides, i t was unstable in the presence of strong base, decomposing to 9-tolualdehyde. with dilute sodium hydroxide solution, the hydroit was easily reduced with bisulfite to p-methyl- peroxide was purified by evaporative distillation benzyl alcohol. Kharasch examined the acid- and re-extraction, a procedure which afforded catalyzed decomposition of I to p-cresol and p - material containing greater than 90% of the theotolualdehyde. 4b Lorand and Edwards observed retical active oxygen. Small amounts of unknown this same decomposition, although the amount of impurities were bothersome and gave erratic phenyl migration differed in the two cases.4c In decomposition results. If the extractions and disaddition, they observed an autocatalytic thermal tillations were carried out with extreme care, decomposition of I, catalyzed by p-toluic acid. however, material of usable purity was produced. No mention was made of the product distribution The identity of all of the 5-7% of impurities present was not known, although 1.5-3% p from this reaction. It was of interest to exarnine this thermal de- tolualdehyde was present. This did not seem to coniposition more closely, in view of the recent affect greatly the decomposition results. demonstration by Mosher5 that primary aliphatic Results and Discussion hydroperoxides decompose to a significant extent The hydroperoxide was decomposed in a helium via a hydroxyperoxide (11) to hydrogen, acid atmosphere a t temperatures of S5-150° as 0.3 AT and aldehyde. The extent to which hydroper- solute in appropriate solvents. Any evolved gas oxide deconiposes in this way depends upon the was collected and measured in a mercury buret. rate of formation of I1 and its equilibrium concen- Prior to analysis, solvent was removed by aqueous tration. For aliphatic systems, this rate is very extraction where possible or by distillation a t fast, and good yields of hydroxyperoxide can be reduced pressure. The resultant products were obtained simply by mixing equivalent amounts then analyzed by three different methods, no one of aldehyde and hydroperoxide. The lower suscep- of which was completely satisfactory for all constituents of the mixture. hilass spectral (h9.S.j (1) C. Walling, “Free Radicals in Solution,” J. Wiley and Sons, Inc., New York, S. Y . , 1957, Chap. 9. analysis was the quickest, but necessitated prior (2) H. S. Blanchard, J. A m . Chem. Sac., 82, 2014 (1960); C. E. H . knowledge of and calibration for all products Bawn, A. A. Pennington and C. F. H. Tipper, Faraday Sac. Disc., 10, of the decomposition. This was not possible 282 (1951); C. E. H. Bawn and D. P. Morgan, J . Inst. Petroleum, 44, generally; first, because all of the minor products 290 (1958); H. Kropf, A n n . , 637, 73, 93 (1960). (3) (a) A . D. Boggs, Ph.D. Thesis, Ohio State University, 1954; could not be identified and, second, because some (b) C. Walling and S. A. Buckler, J . A m . Chem. S a c , 77, 6032 (1955). of the known products exhibited no parent mass (4) (a) H. Hock and S. Lang, Bey., 76B, 169 (1943); (b) M. S. in the spectrum. Gas chromatography was su.tKharasch and J. G. Burt, J . Org. Chem., 16, 150 (1951); (c) E. J. able only for the more volatile coniponents, alcohol, Lorand and E, I. Edwards, J . A m . Chem. Sac., 77, 4035 (1955); W. Pritzkow and R. Hofmann, J . praki. Chem., [41 12, 11 (1960). aldehyde and cresol. The accuracy of both (5) H. S. hlosher and C. F. Wurster, J . A m . Chem. Sac., 77, 5451 methods could be checked by separation of the

+

LA

(1955); C. F. Wurster, L. J. Durham and H. S. Masher, i b i d . , 80, 327 (1958); I,. J. Durham, C. F. Wurster and H. S. Mosher, i b i d . , 80, 332 (1958).

+

+

(6) J. Hine,“Physical Organic Chemistry,” McGraw-Hill Book Cu Inc , h’ew York, N. Y . ,1956, pp 245, 250.

,

&larch 20, 1062

THEDECOMPOSITION OF p-METHYLBENZYL HYDROPEROXIDE

mixture into its major constituents. Acid and cresol were obtained by successive extractions with sodium bicarbonate and sodium hydroxide solutions. Aldehyde was precipitated and determined as its semicarbazone. Chromatography of the residue on alumina was employed for separation and estimation of ester and alcohol. This tedious procedure was used only as a check of the other procedures or when serious discrepancies existed between the M.S. and G.L.P.C. analyses. I n general, deviations from the average of product analyses for duplicate runs were less than 20%. In Table I are shown the products obtained from the decomposition of I in several solvents. The amount of hydrogen produced was not large, amounting to 11 mole per cent. in chlorobenzene solution. Carbon monoxide, carbon dioxide, and, in some cases, oxygen were present in small amounts. I n none of the solvents was the hydrogen yield as great as that obtained for the aliphatic hydroperoxide^.^

1003

=In increased alcohol yield resulted from the decomposition of I in hydrogen-rich solvents such as p-xylene. Abstraction of hydrogen by xylyloxy radicals becomes an important product-determining step in this solvent (3). Oxidation of the resulting alcohol by hydroperoxideg accounts for the increased aldehyde yields also (6). The greater importance of these reactions in p-xylene diminishes the extent of decomposition via the hydroxyperoxide 111. Hence, lower yields of hydrogen (la) and ester (lb) result.

+

RCHiOOH +RCHz0. HO. R’CHj --+ RCHzOH R’CHz. RCHzO. R’CH2. +R‘CHzCHzR’ R’CHz. R’CHI. RCHzOOH +R’CHzOH RCHg0. RCHzOOH + R’CHzOH R’CHO RCHzOH Hz0 K = R’ = CHa-p-C8Hr

+

+

+

+

+

+

+

+

(2) (3) (4) (5) (6)

The p-xylyl radicals produced in eq. 3 may dimerize (4) or attack the hydroperoxide (5).1° To determine that the latter reaction plays a TABLE I significant role in this system, a similar decomSOLVENT EFFECTSos THE DECOMPOSITION OF 0.3 M SOLUposition in p-chlorotoluene solvent was examined. TIOXS O F ~-METHYLBENZYL HYDROPEROXIDE^ In addition to the usual products from I, several -Temp. and solvent 130” 130’ 130’ 130’ 1000 chlorine-containing compounds were observed in Ethylene Acetic pnChlorothe mass spectrum of the product, corresponding Octane Products benzene Xylene glycol acid to p-chlorobenzaldehyde, p-chlorobenzyl alcohol 0.11 0.05 0.09 0.06 0.14 Hz and 4,4‘-dichlorobibenzyl. That these compounds .01 .. .01 .. .. 0 2 contained chlorine was readily apparent from the co .01 0 . 0 1 .003 .. 0.002 ratio of the M and M 2 peaks which agreed .02 .03 ,003 ,. 0.009 coz nicely with the natural abundance of the 35Cl Ester .05 .O1 .03 and 37Cl isotopes. (For the bibenzyl compound, b .22 Acid .08 .27 0.02 the ,21,M 2 and M 4 peaks were used.) Since b .25 Alcohol .46 .32 .15 very little oxygen could have been present in Aldehyde .24 .45 .22 .39 0.3 the system, the oxygenated products must have b p-Cresol .01 .. ,004 resulted from attack of solvent radicals on I (eq. Others 0 . O,sC 0 ,2gd =0.25e 5 and 6, R’ = Cl-p-C6H4-), Expressed in moles per mole of hydroperoxide. Ren-Octane solvent was intermediate to chlorosults not obtained; see text and Experimental. “4,4‘Dimethylbibenzyl. 2-(p-Methylphenyl)-dioxolane. a p - benzene and p-xylene in its effect or1 the decomMethylphenoxymethyl acetate (V). position products. Some increase in alcohol yield was observed, suggesting some hydrogen abstracThe major non-gaseous products of the decom- tion from solvent. In this case, however, only very positions were p-tolualdehyde, p-methylbenzyl small amounts of solvent dimer were detected. alcohol, p-toluic acid and p-methylbenzyl p-toluate. Neither the presence of octenes, from disproporSome @-cresolwas formed also, though usually in tionation of octyl radicals, nor the formation of small amounts. The most rapid decomposition octanols or octanones could be established with occurred in ethylene glycol7 solution with forma- certainty. tion of p-tolualdehyde and its dioxolane. Little The effect of the various products on hydrogen alcohol, acid and hydrogen were produced. Con- formation for the decompositions in chlorobenzene siderably more acid and alcohol were obtained from solution was examined (Tables I1 and 111). the decomposition in chlorobenzene solution, along b‘hereas 9-methylbenzyl alcohol caused a depreswith some ester and p-cresol. These products s:on in hydrogen yield, the others (aldehyde, acid do not afford an adequate oxidation balance for and cresol) caused an increase in hydrogen evoluthe system; the reduction products (alcohol and tion to varying degrees. In contrast to the striking hydrogen) outweigh the observed oxidation prod- differences in decomposition behavior observed ucts (acid, carbon monoxide, carbon dioxide and for aliphatic hydroperoxides in the presence of oxygen) to an appreciable extent. Presumably, their corresponding aldehyde^,^ p-tolualdehyde among the unidentified residuess comprising the exerted only moderate influence on the decompobalance of the reaction mixture may be found suf- sition rate and product composition. Both hyficient oxidized material to account for this dis- drogen and ester yields were increased to some excrepancy. tent. These products become even more prominent a t lower temperatures. Radical traps such (7) C. F. H. Tipper, J . Chem. Soc., 1675 (1953). observed a very

+

+

rapid decomposition of decalin hydroperoxide in this solvent. (8) A number of high mass peaks of low intensity were observed in the m a s spectrum. Some appeared to contain chlorine and may have resulted from radical attack on solvent.

+

(9) M.S. Kharasch, A. Fono and W. Nudenberg, J . Org. Chem., 16, 113 (1951). (10) L.Bateman and H . Hughes, J . Chem. Soc., 4594 (1952); H.E. De La Mare, J. Org. Chem., 25, 2114 (1960).

1004

IYII,LI.%MJ.

as 2,B-di-t-butyl-p-cresol (DBPC) and styrene inhibited hydrogen formation, an effect observed also for I in the absence of added aldehyde. TABLE I1 EFFECTO F ALDEHYDESON THE DECOMPOSITION O F 1' ,_-___ -Conditions------8.5' n-

Products Hi 0 2

1000 p-Tolualdehyde 0.23

co

,003 .01

Ester Acid Alcohol gldehyde Cresol Others

.2 .4 .2

cor

150° $-Tolup-Tolualdehyde aldehyde 0.14